Doses to tumors of up to 80 grays (Gy) have been postulated to eradicate solid experimental tumors with radiommunotargeting,, but this value has proved difficult to reach. Combining two treatment modalities, external beam radiotherapy and radioimmunotargeting, could potentially give rise to a number of advantages.
The purpose of this study was to detect potential benefits with different treatment timing strategies when combining external beam radiotherapy and radioimmunotargeting, with the anticytokeratin monoclonal antibody (MAb) TS1 injected into a nude mouse model carrying subcutaneous human HeLa Hep 2-cell tumors. Cytokeratins are present in necrotic regions within tumors, thereby providing a potential increase in binding sites for TS1 if combined with external beam radiotherapy. External beam radiotherapy was given before, after, and simultaneously with injection of radiolabeled MAb.
The highest yields in terms of total accumulated dose (Gy), percentage of injected activity per gram of tumor tissue, and accumulated dose per injected activity (Gy/MBq) were seen in the group receiving external beam radiotherapy prior to MAb-injection.
Since the pioneering work of Pressman1 in the 1950s, significant progress has been made in the field of radioimmunotargeting (RIT), both experimentally and clinically.2 However, due to low accumulation of monoclonal antibody (MAb) in tumor and bone marrow toxicity as a result of systemic administration of radiolabeled MAb, success with treatment of solid tumors has been delayed.3
Tumor doses of up to 80 grays (Gy) have been postulated to eradicate experimental tumors with RIT, but this level has proved difficult to reach.4 Given the low levels of tumor uptake currently achieved, it appears unlikely that RIT by itself will lead to local control of bulky, solid tumors.5 Using the approach of combining two different treatment modalities—in this case, external beam radiotherapy (EBRT) and RIT—it is possible that the goal of complete eradication of the tumors will be more accessible. With EBRT, radiation is delivered to the tissue at a constant high dose rate. With radiolabeled MAbs, however, radiation is delivered continuously at a low dose rate that increases initially as the MAb accumulates in the tumor, and then decreases due to physical decay of the radionuclide and MAbs clearing from the circulation (biologic half-life). Combining the two modalities could potentially give rise to a number of advantages, a major one being a relative sparing of normal tissues.6 Another possibility is a decrease in interstitial fluid pressure (IFP) as a result of EBRT prior to RIT,7 or an increased accumulation of MAb in the tumor region due to hyperpermeability of the vessels caused by EBRT.8
Several groups have performed combination trials using EBRT and RIT, with varying success. Animal studies have shown that the combination of RIT and EBRT can lead to increased therapeutic efficacy without increasing normal tissue toxicity.9–11 However, in these cases MAbs have been utilized that preferentially target the viable parts of tumors.
In this study, the MAb TS1,12 specific to cytokeratin (CK) 8,13 was used. TS1 is well characterized and has previously been proven useful for experimental RIT.13–16 CKs belong to the intermediate filament family and are major constituents of the cytoskeleton in epithelial cells. In malignancies of epithelial origin, these proteins are deposited in necrotic tumor areas, where they remain for substantial time periods due to their relative resilience against enzymatic degradation. In this investigation, tumors originating from a human cervical adenocarcinoma (HeLa Hep 2) cell line, grown in nude mice, was utilized. Morphometric analysis has demonstrated that approximately 50% of this particular tumor consists of viable tumor cells, the remaining tissue being evenly divided between necrosis and connective tissue.14
Using the TS1-CK8 system, we hypothesized that a favorable effect could be achieved by combining the two treatment modalities, as a result of increased accessibility of antigen following EBRT. When tumors are subjected to EBRT, necrosis of tumor cells follows, leading to an increased deposition of accessible CKs within the tumors. Hence, a larger amount of antigen is available for MAb binding, further potentiating the accretion of radiolabeled MAb.
The purpose of this study was to detect potential benefits with different treatment timing strategies when combining EBRT and RIT with the anti-CK8 MAb TS1.
MATERIALS AND METHODS
The anti-CK8 MAb TS1 is of the immunoglobulin (Ig)G1κ isotype. For this study, it was purchased from InRo Biomedtek (Umeå, Sweden).
Animal Model for Combination Therapy
Twenty-five female nude mice ages 5–8 weeks (Balb/C nu/nu, Bomholtgaard, Denmark) were inoculated with 2 × 106 HeLa Hep 2 cells subcutaneously, centrally on the back, approximately 1 cm from the root of the tail. The cells had been cultured in RPMI 1640 medium (Gibco, Scotland) containing 5% (volume per volume [v/v]) fetal calf serum and 1% (v/v) penicillin, streptomycin, prior to inoculation. HeLa Hep 2 cells originated from a human cervical adenocarcinoma line that expressed CK8, CK18, and CK19.
The MAb TS1 was iodinated with iodine-125 (125I) (IMS 30, Amersham Pharmacia, Uppsala, Sweden) using the chloramine-T method. Free iodine was removed using a Sephadex G50 medium column (Pharmacia, Uppsala, Sweden). TS1 was iodinated to a specific activity of 220–343 MBq/mg and α14 to a specific activity of 197 MBq/mg.
External Beam Radiotherapy
In order to determine the appropriate level of EBRT for the combination therapy, an initial experiment was conducted in which the HeLa Hep 2 cell xenografts were subjected to this treatment regimen only. The tumor bearing mice were subjected to 10 or 20 Gy, divided into 4 fractions given in 4 consecutive days. The therapy X-ray unit STABILIPAN (Siemens) was used as radiation source, with 200 kV and a total half-value layer of about 1 mm copper. The field size was constant with a collimator insert and a constant focus to tumor distance. The dose rate was 1.7 Gy/min and the treatment was given with two opposing X-ray fields against the tumor. To assure a reproducible treatment, preirradation warm-up procedures were performed prior to every treatment occasion. The mice were placed in specially designed lead containers (provided by Prof. Juliana Denekamp, Translational Group, Department of Oncology, Umeå University, Sweden), which facilitated the deposition of irradiation to the subcutaneous tumor and protruding skin only.
The tumor volumes were measured in three dimensions approximately twice a week throughout the study period of 78 days. When the tumor volume had doubled, the mice were excluded from the study and sacrificed.
Combined External Beam Radiotherapy and Radioimmunolocalization
The animals were divided into the Groups A–F. Group A was given 125I-TS1 only, while Group B was given 15 Gy of EBRT prior to 125I-TS1 injection. Group C received EBRT and 125I-TS1 simultaneously, and Group D was given 125I-TS1 prior to the EBRT (Table 1). As a control, Group E was given the nonspecific radiolabeled MAb α14 only, while Group F received EBRT prior to 125I-α14 injection.
All groups were injected with radiolabeled monoclonal antibody on Day 1. Groups A and E were treated with similar doses; but different monoclonal antibodies, as were Groups B and F.
RIL 100 μg
RT 5 Gy
RT 5 Gy
RT 5 Gy
RIL 100 μg
RT 5 Gy
RT 5 Gy
RT 5 Gy
RIL 100 μg
RIL 100 μg
RT 5 Gy
RT 5 Gy
RT 5 Gy
RIL 100 μg
RT 5 Gy
RT 5 Gy
RT 5 Gy
RIL 100 μg
As can be seen in Table 1, the EBRT was given as 3 fractions over 3 consecutive days, i.e., 5 Gy per day and 15 Gy in total, using the same methodology as described above. The MAbs were administered as 100 μg of the 125I-labeled MAb TS1 (Groups A–D), or 100 μg of the 125I-labeled MAb α14, directed against α2-macroglobulin (Groups E–F).
Group A was given 125I-TS1 without receiving any EBRT. Group B was injected with 125I-TS1 2 days after completion of the EBRT, while Group C was given EBRT and 125I-TS1 simultaneously. Group D received the 3-day EBRT treatment 2 days after 125I-TS1 injection. The control group, Group E, was given the nonspecific radiolabeled MAb α14 only, while Group F was given EBRT 2 days prior to 125I-α14 administration, similarly to Group C (Table 1). The MAbs had been diluted in 0.25 M phosphate-buffered saline (pH 7.5), and all mice were injected intraperitoneally. The animals were fed pellets and water ad libitum. The water was supplemented with potassium iodide (Lugol solution; 10 mM NaHCO3 supplemented with 1 mg KI/mL) in order to block thyroid uptake of free iodine.
A scintillation camera (Porta Camera, General Electric, West Milwaukee, WI) connected to a Hermes (NUD, Stockholm, Sweden) evaluation system was used for the radioimmunoscintigraphic measurements (RIS), to determine the distribution of radioactivity in the mice. A holder was used to situate the anesthetized animal at a distance of 10 cm from the pinhole collimator, in order to maintain constant geometric conditions. The mice were anesthetized intraperitoneally prior to RIS using 0.1–0.2 mL of a 1:1:2 cocktail of Dormicum (Midazolam 5 mg/mL; Roche AB, Stockholm, Sweden), Hypnorm (Fluanisonum 10 mg/mL and Fentanylum 0.2 mg/mL; Jansen Pharmaceutica, Beerse, Belgium), and sterile water.
The study period was 48 days in total (35 days for the controls), and each animal was subjected to 23 consecutive RIS measurements. At the end of the study period, each animal was sacrificed and dissected, and the activity in the tumor and each organ was measured. No differences from earlier experiments regarding the biodistributions in organs and blood were obtained (data not presented). Dose calculations were made using Medical Internal Radiation Dose formalism.17
In the initial EBRT experiment, where the animals were subjected to 10 or 20 Gy total radiation doses fractionated over 4 consecutive days, a slight growth inhibition was seen at 10 Gy (Fig. 1). The response increased significantly when the dose was increased to 20 Gy.
As is indicated in Figure 2, which depicts the percentage of injected activity per gram of tumor tissue (% injected activity/g tumor tissue) over the 48-day study period, Group B peaked on Day 5 after MAb injection and had the highest % injected activity/g tumor tissue value of all groups. After Day 5, there was a rapid decrease until approximately Day 10, after which the curve leveled off and ran parallel to the other groups. Group A (125I-TS1 alone) had a lower % injected activity/g tumor tissue value than Group B (125I-TS1, 2 days after final RT) (P = 0.0001) throughout the entire study period. The uptake curve displayed no peak and remained relatively constant until Day 25, after which it initiated a discrete decrease. The % injected activity/g tumor tissue curves for Groups C and D were quite similar to those for Group B, although significantly different (P < 0.05), displaying a resemblance to Group B in shape. They both demonstrated a peak on Day 10, i.e., approximately 5 days after Group B, which was followed by a rapid decrease over the following 10 days, after which they stabilized. After Day 20, the curves for all groups were more or less parallel. At the end of the study period (Day 48), the difference in % injected activity/g tumor tissue between the four groups was insignificant.
In Figure 3, the mean accumulated dose per injected MBq values (Gy/MBq) is displayed for each group. Group B reached a total value of 0.032 Gy/MBq, compared with approximately 0.02 Gy/MBq fo Groups A, C, and D. A difference can be observed in the rate of increase in Gy/MBq, especially between Groups A and B, during the first 10 days of the study. Group B demonstrated a more rapid increase than Group A, and reached a significantly higher total value (P = 0.012). During the latter half of the study period, however, this difference diminished and the curves were almost parallel. Groups C and D initially had a greater rate of increase than Group A but a lower rate than Group B. However, after Day 13, the high rate subsided, and during the remaining period a lower rate, close to the one in Group A, was seen. For the controls, Groups E and F, the values were lower than for all other groups, both in terms of Gy/MBq and the rate of increase.
The total radiation dose to the tumors (Gy) for Groups A-D is visualized in Figure 4. Group B reached the highest dose (0.8 Gy) compared with Groups A (P < 0.05), and C and D (P < 0.05). Similarly to Figure 3, the rate of increase in tumor dose was higher for Group B than for the other groups.
In Figure 5, the decrease in mean total body activity over time (%) is shown. As can be seen, there were no significant differences between the four groups, indicating that there was no difference in the clearing of potentially shedded antigen from the circulation.
The results obtained in this study indicate that an increase in the accumulation of radiolabeled MAb in HeLa Hep 2-cell tumors can be achieved by combining EBRT and RIT.
The choice of a total EBRT dose of 15 Gy for the combined treatment study was based on the results from the EBRT trial (Fig. 1). It was observed that 10 Gy did not cause measurable tumor growth inhibition, while 20 Gy did. As a compromise, 15 Gy was given in 5-Gy fractions over 3 consecutive days, since it appeared probable that a substantial necrosis development could be expected at this dose.
In terms of % injected activity/g tumor tissue (Fig. 2), Gy/MBq (Fig. 3), and radiation dose (Fig. 4), the most favorable combination of treatment was seen for Group B, where EBRT was given prior to injection of 125I-TS1 (Table 1). The finding that Group B both peaked early and generated the highest % injected activity/g tumor tissue (Fig. 2) illustrates that this particular treatment sequence yielded a significantly higher and more rapid 125I-MAb uptake than the other groups. The greater amount of 125I-TS1 per gram of tumor tissue during the first 7 days may be related to the presence of a larger number of released and accessible CKs for MAb binding, caused by an increase in necrotic areas following EBRT. In comparison, the group receiving 125I-TS1 only illustrates no major peak and reached only half the maximum level of Group B. From Day 20, the curves converged and thereafter ran in a parallel fashion. This implies that a positive effect was achieved following the combination of EBRT and RIT that persists up to 20 days, compared with treatment with 125I-TS1 alone.
In Groups C and D the end-results were similar to those for Group B, although the kinetics varied. The shape of the curves for Groups C and D resembles that for Group B, with peaks after the EBRT, although appearing a few days later, similarly followed by a constant decrease. These results are in accordance with our predictions, since it appears reasonable that a peak in MAb accretion should be reached some days after EBRT exposure, when necrosis development has commenced. As seen in Figure 2, the peaks of Groups C and D practically coincided in time, only differing by 1 or 2 days. Group C preceding Group D is in agreement with how the treatments were given, with EBRT for Group C commencing on Day 1 and for Group D on Day 3 (Table 1). All the reported “uptake values” in this investigation (Fig. 2) were low, due to the “background subtraction” for tissues surrounding the tumor. Without this subtraction, all values would have been approximately five times higher. Figures 3 and 4 further support the hypothesis, demonstrating that the largest tumor dose and Gy/MBq was found in Group B, compared with the other groups. These results indicate that EBRT has an effect on MAb uptake in the tumor when this particular system is used. There are a number of potential explanations for this effect, e.g., an increase in the number of potential binding sites for the radiolabeled MAbs in the tumor following EBRT; facilitated diffusion of MAbs due to a decrease in IFP; or a local inflammatory reaction in the tumor, leading to a higher vascular permeability.8, 11 As can be seen in Figure 3, the two control groups, for whom an unspecific MAb was used, generally had lower uptake than groups injected with TS1. Furthermore, the difference between the group injected with 125I-α14 and the group treated with EBRT prior to 125I-α14 injection was not significant. This suggests that putative positive effects are not caused by unspecific mechanisms such as a decrease in IFP or increased vascular permeability.
Sun et al. reported that tumor necrosis or fibrosis reduced the MAb uptake to about 60% after treatment with 30 Gy of EBRT. This was explained by their use of carcinoembryonic antigen as target, a plasma membrane–bound antigen. As regions of necrosis or fibrosis develop following EBRT, a decreased number of potential binding sites is the consequence.11 An advantage of the TS1-CK system is that the number of binding sites increases with the development of necrosis. Furthermore, the possibility of potential synergistic effects with 131I instead of 125I, leading to further enlargement of the necrotic regions, is beneficial.
The difference in uptake of MAbs between the four groups was not due to different clearing rates of MAb from the circulation; none of the groups displayed a more rapid clearance of MAb due to shedded antigen following EBRT treatment, as could have been expected (Fig. 5).
Combining these two modalities gives rise to a number of potential advantages, a major one being the sparing of normal tissues. Buchegger et al. have shown in animal models that it is possible to increase the therapeutic effect without increasing normal tissue toxicity.10
In conclusion, EBRT preceding RIT proved to be the most beneficial treatment sequence in terms of 125I-MAb accumulation when the anti-CK MAb TS1 was used with nude mice xenografted with human HeLa cells. These findings may be of clinical importance in the future, as they offer the potential of increasing the dose to solid tumors.